Automated thermographic screening can be very useful and effective for conducting breast cancer screening where knowledge and expertise of the specific Region of Interest (ROI) of the breast is required to separate a specific region of interest from surrounding tissues which are of no interest so that the identified tissue can be further analyzed and subsequently classified as being cancerous or non-cancerous. Extraction of one or more regions of interest in a thermo graphic image of is a challenging task due to the amorphous nature of the tissue undergoing examination. Although experts can segment the breast tissue into regions of interest by visual inspection after years of experience, it becomes a difficult task to manage effectively when the number of patients being screened is larger than a few. As sophisticated software tools for breast cancer screening arise in the medical software arts, there is an increasing need for methods which can automatically identify and segment regions of interest in a thermographic image of a patient's breast. The teachings hereof are specifically directed to this ongoing effort.
Accordingly, what is needed in this art are increasingly sophisticated systems and methods for automatically segmenting breast tissue from surrounding tissue in a thermal image of a patient for breast cancer screening.